This is a Revision Application. Alterations of serotonin (5-HT) and dopamine (DA) activity may contribute to extremes of appetitive behaviors in anorexia nervosa (AN) and bulimia nervosa (BN), through effects on inhibitory and reward neural pathways. To avoid the confounding effects of malnutrition, and because they have behaviors and neural circuit alterations relevant for this study, we will study 25 recovered (REC) restricting-type AN, 25 REC bulimic-type AN (AN-BN), 25 REC BN, and 25 control women (CW). This 5 year study, of women 18 to 45 years old, will employ positron emission tomography (PET) imaging with radioligands for the 5-HT transporter ([11C]DASB) and DA D2/D3 receptors ([11C]raclopride). Unchanged Aims from the parent application: Aim 1 predicts that REC AN will have elevated, and REC BN will have diminished [ C]DASB binding potential (BPND) in midbrain and striatal regions consistent with 5-HT 11 contributions to enhanced or diminished inhibition. Aim 3 tests hypotheses related to 5-HT and DA interaction contributions to a balance between (respectively) an aversive or inhibitory system, and a reward or motivational system. Aim 4 will integrate PET findings with data from a previously funded application (using the same subjects) in which fMRI characterizes neural processes related to aspects of reward, behavioral inhibition, and appetitive regulation of sucrose. Aim 5 explores relationships between genetic alterations and imaging findings. In this Revision: Disturbances of DA function may contribute to AN and BN being unable to appropriately respond to reward or inhibit appetitive stimuli. In Aim 2 (now 2A), we predict that REC AN will have increased [ C]raclopride BPND in the anterior ventral striatum and REC BN will have reductions in 11 associative striatal regions. This Revision Application adds a PET/double [11C]raclopride/amphetamine (AMPH) paradigm to assess striatal endogenous DA release. Aim 2B predicts that endogenous DA release will be diminished in limbic striatal regions in REC AN, and in associative and sensorimotor striatal regions in REC BN, consistent with diminished ability to direct motivated responses in AN and hypoactive dorsal associative inhibitory pathways in BN. AIM 2C is the primary focus of this application. Because palatable foods stimulate DA secretion, this supplement proposes to use this AMPH paradigm as a probe of AN response to food. AN tend to have a paradoxical response to eating, generating anxiety rather than pleasure. Our data shows measures of anxiety and inhibition in AN are positively associated with dorsal caudate (DC) [11C]raclopride BPND and DA release, consistent with studies implicating DC D2 receptors with risk avoidance and inhibition. Effective treatments for AN and BN have been elusive, and there is little understanding of puzzling symptoms such as extremes of eating behaviors. A better understanding of the underlying neurobiology is critical for being able to develop specific and more powerful therapies for these often chronic and deadly disorders.